Neurophysiological Correlates of Neurological Deficiency in Hemodynamically Significant Stenosis of the Arteries of the Neck
Roman Alexandrovich Zorin, Vladimir Alekseevich Zhadnov, Roman Yevgenyevich Kalinin, Alexander Sergeevich Pshennikov, Nikita Andreevich Solianik, Alexander Olegovich Burshinov,...
445
ABSTRACT
Background: Determination of correlates and predictors of neurological deficit development in hemodynamically significant stenosis of the neck vessels involves not only indicators of systemic and regional blood flow, metabolic characteristics, but also neurophysiological correlates of cognitive functions, autonomic support of activity.
Aim: Identification of the neurophysiological correlates of neurological deficit in hemodynamically significant stenosis of the neck and head vessels for the usage as predictors of this form of pathology.
Methods: 35 patients with hemodynamically significant stenosis of the vessels of the neck and head, identified by ultrasound scanning and Doppler ultrasonography and verified on the basis of selective angiography were investigated. The division of patients into groups was carried out by the method of cluster analysis based on an assessment of the severity of neurological symptoms and the degree of stenosis. The patients underwent spectral analysis and analysis of the coherence function of the electroencephalogram, registration of cognitive evoked potentials P300, and the study of heart rate variability. The selection of significant neurophysiological correlates of neurological deficit in the groups was carried out by the method of artificial neural networks.
Results: Heterogeneity of the group of patients with hemodynamically significant stenosis of the neck vessels in terms of the severity of neurological symptoms was revealed; the neurophysiological indicators that are of the greatest importance in the distribution of patients into groups with different severity of neurological symptoms based on the technology of artificial neural networks have been determined. The characteristics of the cognitive evoked potential were of the greatest importance in solving this problem.
Conclusion: The method of cluster analysis makes it possible to assess the heterogeneity of a group of patients with hemodynamically significant stenosis in terms of the severity of neurological symptoms; at the same time, predictive technologies based on the machine learning methodology help us to distribute patients into groups with different severity of neurological symptoms.
Keywords: hemodynamically significant stenosis of the neck vessels, neurophysiological parameters, cluster analysis, artificial neural networks.